Thomas Nestmeyer

Thomas Nestmeyer, Jean-François Lalonde, Iain Matthews et al.

Relighting is an essential step in realistically transferring objects from a captured image into another environment. For example, authentic telepresence in Augmented Reality requires faces to be displayed and relit consistent with the observer's scene lighting. We investigate end-to-end deep learning architectures that both de-light and relight an image of a human face. Our model decomposes the input image into intrinsic components according to a diffuse physics-based image formation model. We enable non-diffuse effects including cast shadows and specular highlights by predicting a residual correction to the diffuse render. To train and evaluate our model, we collected a portrait database of 21 subjects with various expressions and poses. Each sample is captured in a controlled light stage setup with 32 individual light sources. Our method creates precise and believable relighting results and generalizes to complex illumination conditions and challenging poses, including when the subject is not looking straight at the camera.

Thomas Nestmeyer, Peter V. Gehler

Separating an image into reflectance and shading layers poses a challenge for learning approaches because no large corpus of precise and realistic ground truth decompositions exists. The Intrinsic Images in the Wild~(IIW) dataset provides a sparse set of relative human reflectance judgments, which serves as a standard benchmark for intrinsic images. A number of methods use IIW to learn statistical dependencies between the images and their reflectance layer. Although learning plays an important role for high performance, we show that a standard signal processing technique achieves performance on par with current state-of-the-art. We propose a loss function for CNN learning of dense reflectance predictions. Our results show a simple pixel-wise decision, without any context or prior knowledge, is sufficient to provide a strong baseline on IIW. This sets a competitive baseline which only two other approaches surpass. We then develop a joint bilateral filtering method that implements strong prior knowledge about reflectance constancy. This filtering operation can be applied to any intrinsic image algorithm and we improve several previous results achieving a new state-of-the-art on IIW. Our findings suggest that the effect of learning-based approaches may have been over-estimated so far. Explicit prior knowledge is still at least as important to obtain high performance in intrinsic image decompositions.

Thomas Nestmeyer, Paolo Robuffo Giordano, Heinrich H. Bülthoff et al.

This paper presents a novel decentralized control strategy for a multi-robot system that enables parallel multi-target exploration while ensuring a time-varying connected topology in cluttered 3D environments. Flexible continuous connectivity is guaranteed by building upon a recent connectivity maintenance method, in which limited range, line-of-sight visibility, and collision avoidance are taken into account at the same time. Completeness of the decentralized multi-target exploration algorithm is guaranteed by dynamically assigning the robots with different motion behaviors during the exploration task. One major group is subject to a suitable downscaling of the main traveling force based on the traveling efficiency of the current leader and the direction alignment between traveling and connectivity force. This supports the leader in always reaching its current target and, on a larger time horizon, that the whole team realizes the overall task in finite time. Extensive Monte~Carlo simulations with a group of several quadrotor UAVs show the scalability and effectiveness of the proposed method and experiments validate its practicability.